Indirect evidence implicates a virus in the etiology of Multiple Sclerosis (MS) and more direct evidence shows that viruses can cause other human demyelinating diseases. Coronavirus, mouse hepatitis virus (MHV), strain A59 infection of mice is a good animal model system for the study of virus induced demyelination. After the initial acute encephalitis, virus becomes impossible to detect during the chronic stages of disease while viral nucleic acids may be detected in white matter. Furthermore, MHV-A59 causes a persistent, productive, but non-lytic infection in primary glial cells in culture. The long term goal of this project is to determine the molecular basis for MHV-A59 persistence in the mouse central nervous system (CNS) and to determine the relationship between viral persistence and chronic demyelination. In this proposal we plan to continue to analyze a weakly demyelinating mutant of MHV-A59 (C12), that we have isolated from persistently infected glial cell cultures. We will compare the pathogenesis of wild type MHV-A59 with C12, to understand the differences between these viruses that result in the weakly demyelinating phenotype of the mutant. More specifically we will compare these viruses as far as persistence in the CNS and the ability to induce demyelination and determine whether persistence is necessary for demyelination. The C12 mutant has five amino acids substitutions in the entire genome compared to wild type virus, tow in the spike protein and three in the replicase. We have demonstrated that the weakly demyelinating phenotype correlates with the presence of Q159L in a receptor binding of the spike protein. We will use in vitro assays to ask whether this mutation alters the interaction if virus with several forms of the MHV receptor or with several CNS cell types. We have recently used targeted recombination to introduce each of the two spike mutations, Q159L and H716D, into recombinant viruses. These will be used to determine whether Q159L is sufficient to result in the weakly demyelinating phenotype. More generally, the recombination technology represent an important step forward in the analysis of the pathogenesis of MHV-A59 in allowing us for the first time to introduce specific mutations into the spike protein and determine the relationship between sequence and pathogenesis. In the final portion of this proposal we will use this approach to introduce the spike gene of the more neurovirulent MHV-4 (JHM) into the A59 genome. This will allow us to determine whether the properties of increase virulence and panencephalitis of MHV-4 are due entirely to the spike protein and if so where they map within the protein.